I wish we could make people do this before posting for helpednisley wrote:Now, if we could just convince everybody to use those squares as a confirmation test, we'd be set: until you can print simple squares correctly, you can't print anything else correctly, so don't even try.dramsey wrote:The solution consisted of fixing everything ...
Ask the MakerGear community for assistance...
Custom 3D printing for you or your business -- quote [at] pingring.org
Here's the process I still use:
https://softsolder.com/2016/04/05/thinw ... libration/
You don't need to use my OpenSCAD source code to generate the models, but you must know the dimensions of your models, however you generate them.
Printing one thinwall (single- or maybe triple- thread) square in the middle of the platform, you can set and verify all the key slicing and mechanical parameters. It will require a few iterations, but it's always much faster than trying to diagnose why a complex part has failed.
With those parameters set, print five identical squares (center and corners) to set & verify proper platform alignment. Again, this may require a few iterations, but you're measuring the actual platform alignment under actual printing conditions, not trying to extrapolate from dial indicator readings.
Once you work through the process, it doesn't take much time at all to verify that the whole software + firmware + hardware stack produces the right result. When you change something (or something changes all by itself), a few squares will help identify what's wrong.
You should also use the squares to calibrate the Extrusion Multiplier for each type of filament.